Download Small-scale platform for high-throughput identification of proteins

Small-scale platform for high-throughput identification of proteins suitable for X-ray structure determination
Ronnie O. Frederick, Won Bae Jeon, Frank C. Vojtik, Megan A. Riters, John Kunert, Paul G. Blommel, Fred R. Blattner, Russell L. Wrobel,
Craig A. Bingman, Craig S. Newman, John G. Primm, George N. Phillips Jr., John L. Markley, and Brian G. Fox
University of Wisconsin-Madison, 433 Babcock Drive, Madison, Wisconsin, USA 53706-1549, http://www.uwstructuralgenomics.org
Introduction: Small-scale cell based
expression screening and preliminary
identification of eukaryotic proteins
suitable for structural determination
The Center for Eukaryotic Structural Genomics (CESG) has developed and
currently uses a rapid small-scale high-throughput screen for identifying
positively expressed cloned genes isolated from various eukaryotic sources
such as Arabidopsis thaliana, rice, human, yeast, and mouse. So far, of over
2,500 genes that have been screened for expression on the small-scale
using E. coli host strains Rosetta-2 and B834-pRARE2, we have found that
~64% are successfully expressed at a small-scale (0.4 mL Terrific Broth or
TB). Of ~1,600 cloned genes that were identified to be expressed on a
small-scale, ~71% have been attempted on a large-scale (2 L) using
methionine auxotroph E. coli host, B834-pRARE (that is required for
selenomethionine, 15N, or 13C/15N enrichment). Of these, ~55% have yielded
a soluble fusion protein that was finally suitable for X-ray crystallization
screening trials (see Table 1). Our present screening protocol has been
designed for speed and enables positive expressing targets to be identified
within 3 days (see Figure 1 and 2). We are currently developing and
implementing new protocols that increase the total number successful
soluble targets going through to X-ray crystallization screening and reduce
the numbers of failed targets that occur downstream (see Figure 3). Table 1
shows that there is a significant attrition of target proteins throughout the
current CESG pipeline scheme. We are currently implementing new
protocols and methodologies that enable earlier screening and removal of
unsuitable target proteins at the small-scale expression screening stage.
The changes in the CESG pipeline design required to achieve these
improvements are: cell culture production (using various E. coli host strains)
and growth at a small-scale, micro-scale protein production and purification,
TEV cleavage (of fusion enhancers), protein detection using novel
fluorescent stains, and protein solubility and mass analysis. We have
successfully used these new approaches to pre-screen and purify large
amounts of several different target proteins using small-scale or micro-scale
cell based protein production and purification techniques. These proteins
were then used successfully in X-ray crystallization screens using a
Fluidigm microcircuit system.
Background: Design of expression
plasmid and strategies for producing
soluble target proteins
The eukaryotic genes selected by CESG are fused to an N-terminal (His)ntagged (n=6 or 8) maltose binding protein (MBP which enhances solubility
and expression levels), and a TEV protease cleavage site is located
between the MBP and target protein (just in front of the cloned gene
segment). The transcription of the cloned genes is controlled by a T5 phage
based promoter that is IPTG or lactose inducible. The eukaryotic genes
were amplified using PCR and inserted at the C-terminal end of this (His)ntag-MBP-TEV cassette fusion using a novel site specific recombination
methodology. The architecture of this expression plasmid allows for
flexibility in manipulating any future design modifications of the plasmid. For
example, the (His)n-tag segment has been increased from 6 to 8 histidine
repeats (this has improved downstream protein purification steps) and the
MBP segment can be readily replaced with thioredoxin, glutathione Stransferase (GST), NusA, or other solubility enhancers. In addition, with all
of the cloned genes stored in inert “entry vectors”, it will be easy to re-insert
the genes into any of the future iterations of the expression plasmid.
For small-scale expression screening, we use Rosetta-2, an E. coli host
strain supplemented with seven rare tRNA codons (codons provided by a
plasmid pRARE2). A second E. coli host, B834-pRARE2, is used for largescale 2-liter cell growths. Figures 1 and 2 show the overview CESG
pipeline workflow and small-scale expression screening methods. The
small-scale expression trials are performed in 96-well growth blocks, at
25oC for 24 hours using either 1 mM IPTG for induction or an autoinduction
method developed by Professor Studier (Novagen Overnight Express
Autoinduction SystemTM). Presently, all genes that expressed successfully
on a small-scale were scored “expression +” and recorded in a laboratory
information management system (LIMS) called Sesame. These positively
expressed genes were then used to transform B834-pRARE2 and grown on
a 2-liter large-scale to produce protein for purification and structural studies.
Bioinformatics
Gene selection
LIMS: Sesame
Positive clones
Cloning:
Gateway and PCR
WG# Received
DNA 96-well
Expression
Screen: 96-well
DNA 96-well
PCR plate
B834-RARE2
B834-RARE2
metmet-
E. coli strains
Isotopic label
Small-scale purification
DNA -80oC
96-well plate
SG13009
SG13009
Rosetta-2
Rosetta-2
MG1655
MG1655
MDS-42
MDS-42
New E. coli strains being tested
E. coli strains with reduced and
minimal genomes
No label
No label
SeMet and 13C/15N
Barcode#
Large-scale 2-L cell growth
NMR studies
Protein purification: IMAC
X-ray studies
The cells used for expressing target proteins were Rosetta-2 cells (using
Novagen Overnight Express Autoinduction SystemTM). One 96-well growth
block was processed (16 times 0.5 mL of culture per target protein). For X-ray
crystallization trials ~50-200 micrograms of purified target protein is needed
(10 mg/mL) in a volume range of 5-50 microliters.
For protein purification, we uses an automated AKTA HPLC. Fusion proteins
are initially purified from ~8 ml cell volume by 1st IMAC capture using 1 ml Ni
affinity column (Amersham Biosciences). After desalting, fusion proteins are
subjected to TEV proteolysis. After cleavage, (His)n-MBP fusion tag and
(His)6-TEV protease are separated from target proteins by a 2nd IMAC
capture. Target proteins are desalted against the buffers specified according
the pI values of targets. Target are concentrated to approximately 10 mg/ml by
using Amicon Microcon® YM-3 or YM-10. Final protein concentrations are
determined by BCA protein assay using BSA as a standard protein.
Figure 1. Summary of the high-throughput protein production CESG pipeine.
Workgroups (WG) of 96 individual eukaryotic cloned genes are processed in this system
at one time. Three types of E. coli host strains (Rosetta-2, SG13009 and B834- pRARE)
are used by the large-scale protein production section (LSPP) to produce proteins.
Protein purification processes the cell pastes generated by LSPP using IMAC
methodologies and provides the pure proteins for structural analysis by NMR
spectroscopy and X-ray crystallography.
A
Comparison of the current small-scale
expression screen and the
new small-scale protein expression
and purification platform being
implemented at CESG
Stage in CESG pipeline
Total
Percent %
Genes isolated by PCR
3,238
100
Genes cloned in expression plasmid
2,632
81
Small-scale expression positive
1,691
52
2-L large-scale protein production
1,361
42
Soluble proteins
1181
37
350
11
Protein purification and structural analysis
(His)n-MBP
imidazole
D
3
4
5
Figure 5. SDS-PAGE analysis of purified target proteins produced on the smallscale.
1. ORF 09383 (9.2 µg by BCA)
2. ORF 19610 (23.6 µg by BCA, 11.0 µg by A280 nm)
3. ORF 19812 (32.5 µg by BCA, 16.7 µg by A280 nm)
4. ORF 34198 (6.3 µg by BCA)
5. BSA standard 10 µg
target
We are also using small-scale protein expression screening to evaluate the
likely success of TEV cleavage. This protocol uses cell lysates without
initiating IMAC based protein purification, and offers a rapid testing of the
success rate of TEV cleavage of target proteins. We were able to match ~90%
of the successful TEV cleavages achieved at the large pipeline-scale using
previously assessed target proteins.
E
target
(His)n-MBP
target
2
TEV protease cleavage analysis
of targets at the small-scale
using whole cell lysates
C
desalted
fusion protein
fusion protein
(His)n-MBP
Rosetta-2
Rosetta-2
target
target
DNA: 96-well plate
stored at -20 degree.
36 x cloned genes:
Studier- chemical
defined plates
13
C, 15N &
Selenomethionine labeled
proteins for purification
96 ZYP0.5G culture plates
Figure 4. Chromatogram and SDS-polyacrylamide gel for the small-scale purification
of mouse protein, BC026994. (A) IMAC capture of His-tagged MBP fusion protein,
(B) desalting of fusion protein, (C), (D) TEV proteolysis of fusion tag isolation of target
by subtractive 2nd IMAC, and (E) final desalting of target protein.
B834-pRARE2
B834-pRARE2
Figure 6. SDS-PAGE gel (4-20%) of 36 targets (soluble vs. cut with TEV).
The cleaved target protein can be seen in lanes labeled 8c, 9c, 10c and 11c.
Summary of small-scale purification yield
SESAME: GENIE module
96-well growth block
TB+glycerol media
ORF
1st IMAC
yield after
desaltinga
(mg)
Purity
of
fusion
protein
(%)
Purity of
target
protein
(%)
00605d
2.5
90
95
-
-
0.27
-
09383
3.1
90
95
21
9.2
0.57
193
(33.9)e
14940f
4.8
90
>40
-
-
0.79
-
19610
4.3
80
70g
18
23.6
0.39
198
(50.8)
19812
6.4
90
95
21
32.5
1.86
683
(36.7)
34198
5.7
70h
95
10
13.3
0.65
133
(20.5)
34255i
ca. 2.0j
90
95
-
-
-
-
4 x 24 well SDS-PAGE
Figure 2. Rapid small-scale high-throughput protein expression screening method. On
day 1, Rosetta-2 cells are transformed and grown on 96 culture plates. On day 2,
colonies are picked and transferred into a 96-well growth block (0.4 mL of TB media),
and induced for 24 hours (with 1 mM IPTG) at 25oC. To screen for positive expressing
clones, the E. coli lysates are analyzed by SDS-PAGE gels on day 3. The positive
expressing clones are recorded in Sesame on day 4. Large-scale growths are started
with freshly transformed B834-pRARE2 cells in the following week.
Rosetta-2
Rosetta-2
Final
volume
(µl)
Final
conc.
(µg/ml)
Expected
yieldb
µg
DNA: 96-well plate
stored at -20 degree.
36 x cloned genes:
Studier- chemical
defined plates
13C, 15N &
Selenomethionine labeled
proteins for purification
96 ZYP0.5G plates
B834-pRARE2
B834-pRARE2
aFrom
Final
yieldc
µg (%)
2 ml (8 ml cell culture) of soluble fraction.
96-well block PA0.5G media
SESAME: GENIE module
bEstimated
Mass spec and
X-ray
crystallization
screens
yield was based upon the yield achieved in large-scale purification.
cDetermined
Protein purification:
IMAC + TEV cleavage
by BCA protein assay.
dPurity
4 x 24 well SDS-PAGE
96-well block TB+G
Overnight Express media
of target was greater than 95%, however, protein disappeared during gel
filtration (1st batch) and concentration (2nd batch).
eFinal
Table 1. Summary of expression screening results.
B
1
Figure 3. New scheme for small-scale high-throughput protein expression screening
and protein purification. On day 1, Rosetta-2 cells are transformed and grown on 96
culture plates. On day 2, single colonies are picked and transferred into a 96-well
growth block (0.4 mL of PA0.5G media), and grown for 24 hours at 37oC. To screen
for positive expressing clones, 20 ml of cell culture are transferred into a second
growth block of autoinduction TB+glycerol Novagen Overnight express media. After
growth at 25oC for 24 hours, the cells are harvested and E. coli lysates were analyzed
by SDS-PAGE gels on day 4. IMAC chemistry was used to isolate the positively
expressed proteins. Soluble target proteins are cleaved from their fusion partners
(MBP) and re-purified, and those that were still soluble and well behaved are checked
by mass spec and then sent for X-ray crystallization trials. Large-scale growths (using
freshly transformed B834-pRARE2 cells) are initiated only for those targets that
showed favorable results on the small-scale.
yield/Expected protein yield x 100.
was found in flow-through fraction after 2nd IMAC and aggregated with
uncleaved fusion protein. Purification failed.
Figure 7. Alternative E.coli host strains have been used to test for positive target
protein expression. For example, we have been using an E. coli deletion mutant
strain, MDS42-pRARE2 to screen for expression using Novagen Overnight
ExpressTM Autoinduction System media (supplemented with selenomethionine).
Growth was at 25oC for 40 hours and the final OD at 600 nm reached ~15. Higher
levels of expression have been seen for selected target proteins in the MDS42pRARE2 E. coli strain (deduced genome with extra insertion elements and several
genes deleted).
Summary
We are improving our protein production system to allow earlier screening
of potentially suitable protein targets by expressing and purifying proteins
on the small-scale. We are also developing new methodologies to increase
the total numbers of protein targets that make their way to the final
structural determination stages. These new protocols are been designed
and implemented around a semi-automated and robotically based system,
to increase the total numbers of iterations and permutations of parameters
that can varied and screened.
fTarget
gGreen fluorescent protein. Purified target was not homogenous. Truncation occurred?
Need to analyze MS data carefully.
hMajor contaminant was truncation product that migrates at the size of MBP on SDSPolyacrylamide gel.
Acknowledgements
Cris Cowan at Promega, Madison, Wisconsin, USA.
Professor William Studier at the Brookhaven National Laboratory for
assistance and advice on autoinduction media and auxotrophic E. coli
strain B834.
iPurity
95% after gel filtration but too little amount to continue. Purification stopped after
gel filtration.
jYield
was estimated from the SDS-Polyacrylamide gel.
Dr. David A. Frisch and Scarab Genomics, LLC, Madison, Wisconsin,
USA for assistance with E. coli MG1655 and MDS-42 strains.
CESG is supported by the National Institute of General Medical Sciences through the Protein Structure Initiative (P50 GM 64598).